Gamma Lot QC

Introduction

This experiment is different from the exp-01.rmd which was just a one off experiment to make sure things look good. In this experiment, 6 different gamma lot reagents were sampled from the gamma shipment and then run.

benchling

conluence doc

There are 3 types of standards in the plate 1. Gamma lot standards 2. Beta lot standards
3. Platinum standards

Platinum standards

Each panel has two sets of platinum standards. Here is how each set is divided

Panel 1 - Platinum Set 1 –> CXCL8, FLT3L, MUC16, IL1R2 Platinum Set 2 –> NA(not part of our 7 marker panel)

Panel 2 - Platinum Set 1 –> CEACAM5, WFDC2 Platinum Set 2 –> TNC

Import data from pate using these commands

pate-rnd xmap-collector --include_file_str GammaLot_QC \--out_file_name gamma_lot_qc.csv

## 
ℹ Downloading user_data_frames/gamma_lot_qc.csv

✓ Saved user_data_frames/gamma_lot_qc.csv to /home/ddhillon/proj…

## [1] TRUE

import data locally

## New names:
## * `` -> ...1

Data summary

Name	Piped data
Number of rows	18432
Number of columns	45
_______________________
Column type frequency:
character	12
Date	1
logical	10
numeric	22
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
xponent_id	0	1.00	3	19	0	66	0
sample_type	0	1.00	5	10	0	4	0
well_id	0	1.00	2	3	0	384	0
well_row	0	1.00	1	1	0	16	0
assay	0	1.00	3	7	0	7	0
conc_units	0	1.00	4	5	0	2	0
instrument_serial	5376	0.71	13	13	0	6	0
computer_name	5376	0.71	9	9	0	6	0
file_name	0	1.00	54	54	0	14	0
xponent_comments	13855	0.25	12	12	0	1	0
FN_comments	0	1.00	6	131	0	21	0
panel	0	1.00	7	7	0	2	0

Variable type: Date

skim_variable	n_missing	complete_rate	min	max	median	n_unique
run_date	5376	0.71	2022-03-18	2022-03-22	2022-03-18	2

Variable type: logical

skim_variable	n_missing	complete_rate	mean	count
Control1_control_expected_concentration	18432	0	NaN	:
Control2_control_expected_concentration	18432	0	NaN	:
Control1_control_range_low	18432	0	NaN	:
Control2_control_range_low	18432	0	NaN	:
Control1_control_range_high	18432	0	NaN	:
Control2_control_range_high	18432	0	NaN	:
calc_conc_min_truncated	0	1	0.20	FAL: 14785, TRU: 3647
calc_conc_max_truncated	0	1	0.01	FAL: 18235, TRU: 197
avg_conc_min_truncated	0	1	0.20	FAL: 14708, TRU: 3724
avg_conc_max_truncated	0	1	0.03	FAL: 17938, TRU: 494

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
…1	0	1.00	26190.91	15011.32	264.00	13067.75	26249.50	39166.25	52211.00	▇▇▇▇▇
well_column	0	1.00	12.50	6.92	1.00	6.75	12.50	18.25	24.00	▇▇▆▇▇
dilution_factor	2016	0.89	3.85	6.59	1.00	1.00	1.00	2.00	20.00	▇▁▁▁▁
standard_expected_concentration	16704	0.09	3963.20	8678.03	2.32	42.79	451.41	2738.75	43820.00	▇▁▁▁▁
bead_count	135	0.99	48.73	10.67	0.00	41.00	48.00	55.00	120.00	▁▇▇▁▁
median_mfi	141	0.99	4450.74	12523.13	24.00	59.00	104.00	1283.00	92603.00	▇▁▁▁▁
avg_median_mfi	12	1.00	4420.06	12472.28	40.58	59.42	99.00	1244.75	90050.00	▇▁▁▁▁
pct_cv_median_mfi	12	1.00	7.65	4.21	0.00	4.63	7.27	9.60	34.31	▇▇▁▁▁
net_mfi	141	0.99	4387.18	12523.67	-26.58	1.00	37.58	1223.92	92551.08	▇▁▁▁▁
avg_net_mfi	714	0.96	4529.05	12686.77	-12.17	0.61	47.54	1411.81	89998.08	▇▁▁▁▁
calc_conc	429	0.98	3020.72	11326.15	0.00	6.17	40.20	988.61	159139.94	▇▁▁▁▁
avg_conc	428	0.98	3020.56	11284.23	0.00	5.71	40.75	1003.78	155942.93	▇▁▁▁▁
conc_pct_cv	5754	0.69	9.21	12.32	0.00	2.15	5.08	11.84	110.87	▇▁▁▁▁
pct_recovery	16739	0.09	102.98	14.45	52.52	95.89	101.21	107.35	210.19	▁▇▁▁▁
calc_conc_truncated	429	0.98	2911.67	10766.89	1.66	12.66	34.92	988.61	159139.94	▇▁▁▁▁
avg_conc_truncated	5754	0.69	3801.87	12155.17	2.34	17.23	105.99	2418.72	155942.93	▇▁▁▁▁
calc_conc_standards_min	0	1.00	14.10	11.06	1.66	4.83	12.52	23.59	35.27	▇▆▁▂▃
calc_conc_standards_max	0	1.00	20381.35	17127.67	2493.60	8126.43	16156.41	31531.02	60903.58	▇▆▂▁▃
avg_conc_standards_min	0	1.00	18.51	14.30	2.34	8.14	15.59	28.62	47.38	▇▆▂▁▃
avg_conc_standards_max	0	1.00	16927.98	15707.44	1437.42	2346.91	15480.60	29068.28	46543.16	▇▆▁▂▃
conc_pct_cv_truncated	0	1.00	4.94	1.41	3.27	3.62	4.74	6.19	7.50	▇▁▂▃▁
kit_lot	5376	0.71	1655232.53	0.50	1655232.00	1655232.00	1655233.00	1655233.00	1655233.00	▇▁▁▁▇

why are there 3 plate 1 for panel 2? - DELETE UNTIL FURTHER CLARIFICATION

## # A tibble: 14 × 2
##    file_name                                                  n
##    <chr>                                                  <int>
##  1 20220317_GammaLot_QC_Panel1_Plate1_20220318_134253.csv  1536
##  2 20220317_GammaLot_QC_Panel1_Plate2_20220318_151455.csv  1536
##  3 20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv  1152
##  4 20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv  1152
##  5 20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv  1152
##  6 20220317_GammaLot_QC_Panel2_Plate2_20220318_151818.csv  1152
##  7 20220321_GammaLot_QC_Panel1_Plate3_20220322_132316.csv  1536
##  8 20220321_GammaLot_QC_Panel1_Plate4_20220322_160459.csv  1536
##  9 20220321_GammaLot_QC_Panel2_Plate3_20220322_133423.csv  1152
## 10 20220322_GammaLot_QC_Panel2_Plate4_20220322_155307.csv  1152
## 11 20220329_GammaLot_QC_Panel1_Plate5_20220330_164222.csv  1536
## 12 20220329_GammaLot_QC_Panel2_Plate5_20220330_170254.csv  1152
## 13 20220330_GammaLot_QC_Panel1_Plate6_20220331_142218.csv  1536
## 14 20220330_GammaLot_QC_Panel2_Plate6_20220331_142301.csv  1152

## # A tibble: 4 × 2
##   file_name                                                  n
##   <chr>                                                  <int>
## 1 20220317_GammaLot_QC_Panel1_Plate1_20220318_134253.csv  1536
## 2 20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv  1152
## 3 20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv  1152
## 4 20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv  1152

Data summary

Name	Piped data
Number of rows	3456
Number of columns	45
_______________________
Column type frequency:
character	11
Date	1
logical	10
numeric	22
________________________
Group variables	file_name

Variable type: character

skim_variable	file_name	n_missing	complete_rate	min	max	empty	n_unique	whitespace
xponent_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	3	19	0	51	0
xponent_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	3	19	0	51	0
xponent_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	3	19	0	51	0
sample_type	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	5	10	0	4	0
sample_type	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	5	10	0	4	0
sample_type	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	5	10	0	4	0
well_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	2	3	0	384	0
well_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	2	3	0	384	0
well_id	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	2	3	0	384	0
well_row	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	1	1	0	16	0
well_row	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	1	1	0	16	0
well_row	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	1	1	0	16	0
assay	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	3	7	0	3	0
assay	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	3	7	0	3	0
assay	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	3	7	0	3	0
conc_units	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	5	5	0	1	0
conc_units	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	5	5	0	1	0
conc_units	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	5	5	0	1	0
instrument_serial	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	13	13	0	1	0
instrument_serial	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	13	13	0	1	0
instrument_serial	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	13	13	0	1	0
computer_name	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	9	9	0	1	0
computer_name	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	9	9	0	1	0
computer_name	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	9	9	0	1	0
xponent_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	967	0.16	12	12	0	1	0
xponent_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	959	0.17	12	12	0	1	0
xponent_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	960	0.17	12	12	0	1	0
FN_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	6	131	0	11	0
FN_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	6	129	0	11	0
FN_comments	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	6	77	0	7	0
panel	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	7	7	0	1	0
panel	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	7	7	0	1	0
panel	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	7	7	0	1	0

Variable type: Date

skim_variable	file_name	n_missing	complete_rate	min	max	median	n_unique
run_date	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1	2022-03-18	2022-03-18	2022-03-18	1
run_date	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1	2022-03-18	2022-03-18	2022-03-18	1
run_date	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1	2022-03-18	2022-03-18	2022-03-18	1

Variable type: logical

skim_variable	file_name	n_missing	complete_rate	mean	count
Control1_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control1_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control1_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
Control2_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control2_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control2_control_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
Control1_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control1_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control1_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
Control2_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control2_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control2_control_range_low	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
Control1_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control1_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control1_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
Control2_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1152	0	NaN	:
Control2_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1152	0	NaN	:
Control2_control_range_high	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1152	0	NaN	:
calc_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1	0.12	FAL: 1019, TRU: 133
calc_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1	0.12	FAL: 1011, TRU: 141
calc_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1	0.12	FAL: 1011, TRU: 141
calc_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1	0.01	FAL: 1143, TRU: 9
calc_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1	0.01	FAL: 1143, TRU: 9
calc_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1	0.01	FAL: 1143, TRU: 9
avg_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1	0.07	FAL: 1072, TRU: 80
avg_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1	0.19	FAL: 928, TRU: 224
avg_conc_min_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1	0.20	FAL: 922, TRU: 230
avg_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1	0.03	FAL: 1118, TRU: 34
avg_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1	0.03	FAL: 1118, TRU: 34
avg_conc_max_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1	0.03	FAL: 1118, TRU: 34

Variable type: numeric

skim_variable	file_name	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
…1	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	10727.09	910.22	9216.00	9899.75	10747.50	11510.25	12275.00	▇▇▇▇▇
…1	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	13799.09	910.22	12288.00	12971.75	13819.50	14582.25	15347.00	▇▇▇▇▇
…1	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	16871.09	910.22	15360.00	16043.75	16891.50	17654.25	18419.00	▇▇▇▇▇
well_column	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	12.50	6.93	1.00	6.75	12.50	18.25	24.00	▇▇▆▇▇
well_column	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	12.50	6.93	1.00	6.75	12.50	18.25	24.00	▇▇▆▇▇
well_column	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	12.50	6.93	1.00	6.75	12.50	18.25	24.00	▇▇▆▇▇
dilution_factor	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	126	0.89	6.33	8.54	1.00	1.00	1.00	20.00	20.00	▇▁▁▁▃
dilution_factor	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	126	0.89	6.33	8.54	1.00	1.00	1.00	20.00	20.00	▇▁▁▁▃
dilution_factor	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	126	0.89	6.33	8.54	1.00	1.00	1.00	20.00	20.00	▇▁▁▁▃
standard_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1044	0.09	5003.22	10390.67	8.11	60.31	601.72	3860.00	43820.00	▇▁▁▁▁
standard_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1044	0.09	5003.22	10390.67	8.11	60.31	601.72	3860.00	43820.00	▇▁▁▁▁
standard_expected_concentration	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1044	0.09	5003.22	10390.67	8.11	60.31	601.72	3860.00	43820.00	▇▁▁▁▁
bead_count	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	135	0.88	44.66	8.56	0.00	38.00	43.00	50.00	85.00	▁▁▇▂▁
bead_count	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	44.64	8.61	0.00	38.00	43.00	50.00	85.00	▁▁▇▂▁
bead_count	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	44.74	8.32	35.00	38.00	43.00	50.00	85.00	▇▅▂▁▁
median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	138	0.88	5572.40	14700.80	33.00	72.00	191.00	2170.00	84573.00	▇▁▁▁▁
median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	3	1.00	4932.00	13920.93	28.00	66.00	118.00	1523.00	84573.00	▇▁▁▁▁
median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	4919.27	13905.01	28.00	66.00	117.25	1484.00	84573.00	▇▁▁▁▁
avg_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	12	0.99	4964.20	13964.66	43.92	65.71	114.08	1520.33	82082.25	▇▁▁▁▁
avg_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	4919.33	13898.78	43.92	64.32	115.38	1520.33	82082.25	▇▁▁▁▁
avg_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	4919.27	13898.80	43.92	64.28	115.38	1520.33	82082.25	▇▁▁▁▁
pct_cv_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	12	0.99	7.70	4.23	0.00	4.47	7.45	10.55	19.42	▅▇▅▃▁
pct_cv_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	8.06	4.35	0.85	4.47	8.42	11.83	19.42	▆▆▇▃▁
pct_cv_median_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	8.07	4.34	0.85	4.47	8.42	11.82	19.42	▆▆▇▃▁
net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	138	0.88	5503.56	14702.67	-18.50	3.42	100.92	2096.56	84506.00	▇▁▁▁▁
net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	3	1.00	4863.17	13922.66	-23.00	0.58	47.92	1451.42	84506.00	▇▁▁▁▁
net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	4850.44	13906.73	-26.58	0.58	46.58	1429.04	84506.00	▇▁▁▁▁
avg_net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	444	0.61	7881.85	17049.31	-2.62	70.92	499.50	3774.50	82015.25	▇▁▁▁▁
avg_net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	18	0.98	4927.49	13996.94	-6.50	-1.32	59.17	1476.42	82015.25	▇▁▁▁▁
avg_net_mfi	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	18	0.98	4927.43	13996.96	-12.17	-1.42	59.17	1476.42	82015.25	▇▁▁▁▁
calc_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	156	0.86	5811.03	17061.95	0.04	15.08	63.75	3988.81	154944.69	▇▁▁▁▁
calc_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	21	0.98	5158.61	16115.24	0.04	15.08	42.79	3069.94	154944.69	▇▁▁▁▁
calc_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	18	0.98	5145.03	16096.05	0.04	15.08	42.79	3054.20	154944.69	▇▁▁▁▁
avg_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	156	0.86	5811.03	16992.95	0.04	15.08	64.56	4001.05	147276.00	▇▁▁▁▁
avg_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	20	0.98	5154.10	16044.48	0.04	15.08	42.79	3065.04	147276.00	▇▁▁▁▁
avg_conc	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	18	0.98	5145.03	16031.76	0.04	15.08	42.79	3065.04	147276.00	▇▁▁▁▁
conc_pct_cv	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	514	0.55	7.51	8.95	0.00	2.53	4.70	8.12	57.92	▇▂▁▁▁
conc_pct_cv	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	358	0.69	6.13	8.53	0.00	0.00	3.99	6.90	57.92	▇▁▁▁▁
conc_pct_cv	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	352	0.69	6.08	8.52	0.00	0.00	3.90	6.80	57.92	▇▁▁▁▁
pct_recovery	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	1048	0.09	102.36	12.98	68.99	96.01	101.73	105.32	141.89	▁▃▇▁▁
pct_recovery	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	1048	0.09	102.36	12.98	68.99	96.01	101.73	105.32	141.89	▁▃▇▁▁
pct_recovery	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	1048	0.09	102.36	12.98	68.99	96.01	101.73	105.32	141.89	▁▃▇▁▁
calc_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	156	0.86	5513.89	15896.96	5.71	15.08	61.01	3988.81	154944.69	▇▁▁▁▁
calc_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	21	0.98	4896.98	15018.61	5.71	15.08	42.79	3069.94	154944.69	▇▁▁▁▁
calc_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	18	0.98	4884.08	15000.80	5.71	15.08	42.79	3054.20	154944.69	▇▁▁▁▁
avg_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	514	0.55	7843.93	18705.37	10.32	60.78	1321.37	7054.25	147276.00	▇▁▁▁▁
avg_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	358	0.69	6366.22	17038.82	10.32	45.65	511.03	4809.97	147276.00	▇▁▁▁▁
avg_conc_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	352	0.69	6318.56	16983.59	10.32	45.65	511.03	4624.72	147276.00	▇▁▁▁▁
calc_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	17.21	11.78	5.71	5.71	12.52	33.39	33.39	▇▇▁▁▇
calc_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	17.21	11.78	5.71	5.71	12.52	33.39	33.39	▇▇▁▁▇
calc_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	17.21	11.78	5.71	5.71	12.52	33.39	33.39	▇▇▁▁▇
calc_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	25228.31	18744.17	8300.00	8300.00	16039.35	51345.58	51345.58	▇▁▁▁▃
calc_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	25228.31	18744.17	8300.00	8300.00	16039.35	51345.58	51345.58	▇▁▁▁▃
calc_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	25228.31	18744.17	8300.00	8300.00	16039.35	51345.58	51345.58	▇▁▁▁▃
avg_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	23.78	15.61	10.32	10.32	15.37	45.65	45.65	▇▁▁▁▃
avg_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	23.78	15.61	10.32	10.32	15.37	45.65	45.65	▇▁▁▁▃
avg_conc_standards_min	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	23.78	15.61	10.32	10.32	15.37	45.65	45.65	▇▁▁▁▃
avg_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	21081.94	18183.14	2157.55	2157.55	15480.60	45607.67	45607.67	▇▇▁▁▇
avg_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	21081.94	18183.14	2157.55	2157.55	15480.60	45607.67	45607.67	▇▇▁▁▇
avg_conc_standards_max	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	21081.94	18183.14	2157.55	2157.55	15480.60	45607.67	45607.67	▇▇▁▁▇
conc_pct_cv_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	3.42	0.00	3.42	3.42	3.42	3.42	3.42	▁▁▇▁▁
conc_pct_cv_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	3.61	0.00	3.61	3.61	3.61	3.61	3.61	▁▁▇▁▁
conc_pct_cv_truncated	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	3.62	0.00	3.62	3.62	3.62	3.62	3.62	▁▁▇▁▁
kit_lot	20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv	0	1.00	1655233.00	0.00	1655233.00	1655233.00	1655233.00	1655233.00	1655233.00	▁▁▇▁▁
kit_lot	20220317_GammaLot_QC_Panel2_Plate1_20220318_141840.csv	0	1.00	1655233.00	0.00	1655233.00	1655233.00	1655233.00	1655233.00	1655233.00	▁▁▇▁▁
kit_lot	20220317_GammaLot_QC_Panel2_Plate1_20220318_143348.csv	0	1.00	1655233.00	0.00	1655233.00	1655233.00	1655233.00	1655233.00	1655233.00	▁▁▇▁▁

Seems like 20220317_GammaLot_QC_Panel2_Plate1_20220318_133136.csv has some missing bead counts - on checking with Jinesh and wilson, they suggest not using this and using the latest run of these 3.

## # A tibble: 135 × 45
##     ...1 xponent_id   sample_type well_id well_row well_column
##    <dbl> <chr>        <chr>       <chr>   <chr>          <dbl>
##  1  9311 Assay Buffer sample      O4      O                  4
##  2  9312 Assay Buffer sample      O5      O                  5
##  3  9313 Assay Buffer sample      O6      O                  6
##  4  9314 Assay Buffer sample      O7      O                  7
##  5  9315 Assay Buffer sample      O8      O                  8
##  6  9316 Assay Buffer sample      O11     O                 11
##  7  9317 Assay Buffer sample      O12     O                 12
##  8  9318 Assay Buffer sample      O13     O                 13
##  9  9319 Assay Buffer sample      O14     O                 14
## 10  9320 Assay Buffer sample      O15     O                 15
## # … with 125 more rows, and 39 more variables: assay <chr>,
## #   dilution_factor <dbl>, conc_units <chr>,
## #   standard_expected_concentration <dbl>, bead_count <dbl>,
## #   median_mfi <dbl>, avg_median_mfi <dbl>,
## #   pct_cv_median_mfi <dbl>, net_mfi <dbl>, avg_net_mfi <dbl>,
## #   calc_conc <dbl>, avg_conc <dbl>, conc_pct_cv <dbl>,
## #   pct_recovery <dbl>, …

Which samples failed bead count? several bead count failures for standard 6 in the platinum set - but none of these contain the standards of interest

1. Remove the wells with bead counts < 35 and create a ‘clean’ data set and output
2. Don’t use the plate 6 for panel 1

Part 1

For the curves fit using gamma lot, we will check the interpolated beta and platinum standard distribution

Outlier test on the results of the 6 lots

Since we have one observation of each beta and platinum standard per lot, we can’t run any statistical test of comparisons. Instead, we want to check if something is wildly different among the 6 lots.

We can achieve this by using the modified z-score outlier test on these 4 observations to detect if something is ‘different’.

Beta and platinum standards

Here we fit the curve using the gamma standards, and then interpolate the beta and platinum standards. We know that these standards are the same sample, simply run with a different lot of gamma reagents and standards.

Their distribution should be within the natural variation range. But since we don’t have any data on these gamma lot reagents, we don’t know how to estimate this natural variation.

Instead, another way to check if they are ‘similar’ is using the modified z score. This is used as an outlier test and from the reagent lots sampled, neither of these should be

Beta Standards

## `summarise()` has grouped output by 'xponent_id', 'file_name'.
## You can override using the `.groups` argument.

## Warning: Removed 6 rows containing missing values (geom_point).

Platinum Standards Set 1

## `summarise()` has grouped output by 'xponent_id', 'file_name'.
## You can override using the `.groups` argument.

CEA, std 5?

## `summarise()` has grouped output by 'xponent_id', 'file_name',
## 'assay'. You can override using the `.groups` argument.

## # A tibble: 6 × 7
##   xponent_id   file_name    assay calc_conc net_mfi computer_name
##   <chr>        <chr>        <chr>     <dbl>   <dbl> <chr>        
## 1 PltSet1 Std5 20220317_Ga… CEAC…      65.0    243. FM3D-CL06    
## 2 PltSet1 Std5 20220317_Ga… CEAC…      64.1    260. FM3D-CL03    
## 3 PltSet1 Std5 20220321_Ga… CEAC…      60.9    207. FM3D-CL01    
## 4 PltSet1 Std5 20220322_Ga… CEAC…      63.8    230. FM3D-CL03    
## 5 PltSet1 Std5 20220329_Ga… CEAC…      65.1    237. <NA>         
## 6 PltSet1 Std5 20220330_Ga… CEAC…      68.2    243. <NA>         
## # … with 1 more variable: mean_bead_count <dbl>

Platinum Standards Set 2

## `summarise()` has grouped output by 'xponent_id', 'file_name'.
## You can override using the `.groups` argument.

## Warning: `gather_()` was deprecated in tidyr 1.2.0.
## Please use `gather()` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.

What’s going on for TNC std 6? (plate 3 and plate 5 seem very different)

## `summarise()` has grouped output by 'xponent_id', 'file_name',
## 'assay'. You can override using the `.groups` argument.

## # A tibble: 6 × 7
##   xponent_id   file_name    assay calc_conc net_mfi computer_name
##   <chr>        <chr>        <chr>     <dbl>   <dbl> <chr>        
## 1 PltSet2 Std6 20220317_Ga… TNC        21.2    51.3 FM3D-CL06    
## 2 PltSet2 Std6 20220317_Ga… TNC        21.8    51.9 FM3D-CL03    
## 3 PltSet2 Std6 20220321_Ga… TNC        24.9    54.6 FM3D-CL01    
## 4 PltSet2 Std6 20220322_Ga… TNC        21.4    55.3 FM3D-CL03    
## 5 PltSet2 Std6 20220329_Ga… TNC        18.5    50.3 <NA>         
## 6 PltSet2 Std6 20220330_Ga… TNC        21.9    73.7 <NA>         
## # … with 1 more variable: mean_bead_count <dbl>

Between run CVs

Platinum Standards Set 1 and 2

## `summarise()` has grouped output by 'file_name', 'assay'. You
## can override using the `.groups` argument.
## `summarise()` has grouped output by 'assay'. You can override
## using the `.groups` argument.

## `summarise()` has grouped output by 'file_name', 'assay'. You
## can override using the `.groups` argument.
## `summarise()` has grouped output by 'assay'. You can override
## using the `.groups` argument.

Generally speaking, CVs are higher for Std 6 and Std1 for a lot of the markers, so it isn’t surprising to see some outliers in the platinum standards for S6 and S1. If there were true differences in one of the lots, we would have seen this consistently.

samples

Part 2

In order for us to value assign the gamma lot standards against our platinum standards, we will

1. Fit a curve using the platinum standards for each of the 6 runs

2. Interpolate the gamma lot standards to the curve fit by platinum standards

3. Take the average across all 6 runs and assign that as the value of each gamma lot standard

Create Platinum Standard Data

We need to send the value assignments to RDS for ALL 20 proteins, not just the 7 markers because RDS is not aware that we are only going forward with the 20.

## New names:
## * `` -> ...1

Value assignment of platinum standards

This deck provides with the value assignments

Create a dataframe for fitting the curve

Now we have the data we need to fit a curve using the platinum standards. The next thing we need is the gamma lot standards data which we will “infer” on

## # A tibble: 648 × 4
##    xponent_id   file_name                             assay     n
##    <chr>        <chr>                                 <chr> <int>
##  1 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… AFP       6
##  2 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… CCL20     6
##  3 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… CXCL8     6
##  4 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… EGF       6
##  5 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… FLT3…     6
##  6 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… HGF       6
##  7 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… IL1R2     6
##  8 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… IL6       6
##  9 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… KLK3      6
## 10 PltSet1 Std1 20220317_GammaLot_QC_Panel1_Plate1_2… MUC16     6
## # … with 638 more rows

Scipy Functions

We also define a function that uses the average over the replicates and then fits the curve.

Fit curve and infer data

Data summary

Name	Piped data
Number of rows	648
Number of columns	9
_______________________
Column type frequency:
character	4
list	1
numeric	4
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
assay	0	1	3	7	0	20	0
xponent_id	0	1	9	9	0	6	0
sample_type	0	1	6	6	0	1	0
file_name	0	1	54	54	0	11	0

Variable type: list

skim_variable	n_missing	complete_rate	n_unique	min_length	max_length
units	648	0	0	1	1

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
net_mfi	0	1.00	14163.02	21739.15	11.75	255.52	2429.58	20020.27	107303.9	▇▂▁▁▁
standard_expected_concentration	0	1.00	8773.47	27161.76	0.04	57.97	504.20	3985.00	234160.0	▇▁▁▁▁
bead_count	0	1.00	47.72	9.91	35.00	40.00	47.00	54.00	81.0	▇▆▃▁▁
inferred_concentration	13	0.98	45885.47	273323.46	0.00	22.35	513.96	6921.13	3134051.2	▇▁▁▁▁

Several missing inferred_concentrations for the standard 1 in gamma lot (MIF, OPN, MUC1)

net_mfi for S1 for MUC1 gamma (and beta) is way higher than platinum standard - for the gamma lot S1 that are outside the S1 , we will assign them the S1 from the platinum standard itself since they are not in the quantiation range.

## # A tibble: 13 × 9
##    assay xponent_id net_mfi sample_type standard_expected_… units
##    <chr> <chr>        <dbl> <chr>                     <dbl> <lis>
##  1 MUC1  Standard1   43368. sample                       40 <dbl>
##  2 OPN   Standard1   35532. sample                   234160 <dbl>
##  3 MIF   Standard1   40445. sample                   129470 <dbl>
##  4 MIF   Standard1   46497. sample                   129470 <dbl>
##  5 MUC1  Standard1   37998. sample                       40 <dbl>
##  6 MUC1  Standard1   42132. sample                       40 <dbl>
##  7 MIF   Standard1   37366. sample                   129470 <dbl>
##  8 MIF   Standard1   44298. sample                   129470 <dbl>
##  9 CCL20 Standard1   45934. sample                     6990 <dbl>
## 10 MUC1  Standard1   38321. sample                       40 <dbl>
## 11 OPN   Standard1   31437. sample                   234160 <dbl>
## 12 MIF   Standard1   39026. sample                   129470 <dbl>
## 13 MIF   Standard1   42978. sample                   129470 <dbl>
## # … with 3 more variables: bead_count <dbl>, file_name <chr>,
## #   inferred_concentration <dbl>

## Warning in self$trans$transform(x): NaNs produced

## Warning: Transformation introduced infinite values in continuous
## x-axis

## Warning: Removed 2 rows containing missing values (geom_point).

Let’s generate these plots for all gamma standards for ALL proteins

## [[1]]

## 
## [[2]]

## 
## [[3]]

## 
## [[4]]

## 
## [[5]]

## 
## [[6]]

## 
## [[7]]

## 
## [[8]]

## 
## [[9]]

## 
## [[10]]

## 
## [[11]]

## 
## [[12]]

## 
## [[13]]

## 
## [[14]]

## Warning in self$trans$transform(x): NaNs produced

## Warning: Transformation introduced infinite values in continuous
## x-axis

## Warning: Removed 2 rows containing missing values (geom_point).

## 
## [[15]]

## 
## [[16]]

## 
## [[17]]

## 
## [[18]]

## 
## [[19]]

## 
## [[20]]

Final Assignments

For the final assignments, we will take the average over all the runs (5 for panel 1 and 6 for panel 2), and for the ones that are beyond the quantitation range of S1, we simply assign them S1 (even if they are quantifiable because we are not confident in our extrapolations).

## `summarise()` has grouped output by 'assay'. You can override
## using the `.groups` argument.
## Joining, by = "assay"
## Auto-refreshing stale OAuth token.
## x Request failed [429]. Retry 1 happens in 2.4 seconds ...
## x Request failed [429]. Retry 2 happens in 7.9 seconds ...
## x Request failed [429]. Retry 3 happens in 25.7 seconds ...
## ✓ Writing to "gamma-lot-specs-2022".
## ✓ Writing to sheet 'value_assignments'.

Part 3

We need to generate some estimate of precision for the samples from gamma lot (for all markers - not just the 7 proteins)

We will do this for the clinical reference samples only that we are sending to RDS.

1. We subset the reference samples and the gamma standard data only
2. We will change the standard_expected_concentration to the value assignments we did in part 2.
3. We will proceed with fitting the curve and then retreive the inferred_concentrations for the reference samples.

Join with the gamma assignments

Intra-CVs

We will provide intra-CVs for

1. Concentrations of the samples

## `summarise()` has grouped output by 'xponent_id', 'assay'. You
## can override using the `.groups` argument.
## `summarise()` has grouped output by 'xponent_id'. You can
## override using the `.groups` argument.
## `summarise()` has grouped output by 'xponent_id', 'assay'. You
## can override using the `.groups` argument.
## `summarise()` has grouped output by 'xponent_id'. You can
## override using the `.groups` argument.

## ✓ Writing to "gamma-lot-specs-2022".

## ✓ Writing to sheet 'Reference Samples Specs'.

2. Standards (do we need to provide specs on standards?)